The study on animal husbandry social ecosystem analysis and multi-agent model based on family farm in Qilian mountains

The Qilian Mountains Project resulted from the collaboration of ICTPEM (International Center for Tibetan Plateau Ecosystem Management, Lanzhou University, Gansu Province, China) and Agreenium (Cirad, Inra and AgroParisTech) started in 2007 during the preparation of 8th IRC/IGC (International Rangeland & Grassland Congress (Hohhot, Inner Mongolia, July 2008). Rangeland area located in the highlands of the North-Eastern Tibetan plateau. Animal breeding, especially yaks and sheep, is the main agricultural activity due to the harsh climatic conditions. As all the Tibetan Plateau, the great challenge in the Qilian Mountains is to maintain a rural activity for Chinese minorities without degraded the rangeland which is the water tower of urban and rural industrialized China. The main objective of the Qilian Mountains Project is to describe, understand and model the farming systems in the Qilian Mountains, then to assess their resilience and to draft some scenarios for the policymakers. Another objective is capacity building at post-graduate level, especially in interdisciplinary and integrated research. Develop Chinese-French partnership is also another strategic objective. Until now, three groups of research actions have been developed. First, in 2012, just after the project approval, a survey based on interviews with 25 local stakeholders (breeders, local leaders, traders, input providers, local policymakers, extension services, etc.) aimed to draft the long-term history of rural activity in the zone and identify the hopes and fears of local people regarding the future. Secondly, in 2013, another survey based also on interviews with local stakeholders allowed to better understand the main drivers of local trends, rural activities and the major constraints of farming systems in two contrasted areas, the Tianzhu and Sunan county. In 2014, a survey based on questionnaires with 50 farms in 8 villages, 4 in each county allowed to describe the different farming systems with qualitative and quantitative data. In the same time, two specific researches have been launched, one to assess the households' economy and another focused on the supply-market chains. Two participative workshops to present and debate the results, one in each county, allowed to better define the local demand in terms of research, research-development and training. Two research actions are planned for 2015, one on the rangeland/pasture management using MODIS images along the 25-30 past years in partnership with ESO Unit Research, another on supply chain management in partnership with AgroParisTech